Rough bank on a v8

[TheShagg]

Engine: ford 5.0 HO
ignition: EDIS 8
injectors: 35lb low impedence, using flyback board and 30% PWM setting.

Looking for some ideas from people who know what they are doing!

This is a new build, so I can't say the engine has ever run right to begin with.

I can get the engine to idle, but it is a touch rough. When I put my hands over both exhaust pipes (no crossover in the exhaust) the driver side has a nice even *put put put put* but the passenger side is out of rythm and its like "put put put PUT put PUT PUT put". Also, after a lot of idling some of the header primaries on the passenger side will start to glow.

Using my wideband I see a good 14.7-ish stoich ration. It bounces around a lot between 13's and 15's. This is on the driver side (good bank). I have not yet tried the passenger side. Unfortunately my Innovative LC-1 blew out the analog channels and I am in the middle of trying to RMA the thing, but they are dragging their feet.

If I boost up the fuel table quite a bit, so that the AFR reads 12's-13's, (boosted the VE map from say 55% to 70%+) the engine smooths out, picks up RPM, the header stops glowing, the exhaust pulses feel much nicer, and my eyes start to sting.

1) I tried swapping the injector harness connections between the left and right banks (I have each bank on a seperate harness, INJ1 and INJ2), i.e. i swapped them between the INJ1 and INJ2 outputs. No change.
2) I tried swapping out a couple of injectors on the most suspect cylinders (based on header tube glow) and there was no change. I will try swapping the injectors across banks soon.
3) I swapped the 2 edis coils with eachother - no change.

My thoughts are that it's either:
injectors: doubtful because of my previous swapping, but I will do a full bank-for-bank swap soon.
ignition: perhaps one of the 4 coil drivers is messed up. Not sure why adding fuel would help this except maybe its rpm related.
leaking intake gasket: maybe it really is lean?
valves: I have no idea on this one.... if nothing else fixes it, then this is probably the problem?

Thanks for any help!

[LT401Vette]

might be silly, but any chance a couple plug wires are crossed on the passenger side?

[TheShagg]

Alright, so I messed with moving around injectors, and the results are inconclusive, will mess with it more tomorrow. Apparently I am losing faith in my "feel exhaust pulses with hands" method. As the car is right now I can't figure out if one side is worse than the other, so I will rewind one step and see if there is a change. Need to wait for some new daylight.

However, I hooked up my LC1 and used the digital output, and found some interesting results. According to both banks, at the moment, I was running lean - about 16 AFR. This is on both banks. Perhaps my analog measurements I was using from MegaTune were simply incorrect, due to ground offset - my ground wire was most likely floating while I made all the observations in the first post, which I think is why they analog outputs blew out.

Also both banks seem to have an OK agreement on the AFR - remember I only have one LC1 so I have to shut off the engine between runs. And it bounces around a lot so it's hard to tell.

I richened it up, and it still feels not quite as smooth as a sewing machine. My hand-o-meters can't make heads or tails of that at the moment.

Here is a graph at idle. Does this look like its bouncing around too much?

AFR1.png

[trakkies]

Most older engines will need something like 13.5 - 14:1 for best hot idle. And IIRC, a 90 degree V8 will never give a perfect idle as the firing pulses are irregular. For that you need a 6.

Again IIRC, most V8s link the two sides of the exhaust together to get the best extraction effect. To do this with two separate ones would mean crossing over exit pipes between banks.

[TheShagg]

Yep, I don't have any crazy 360 degree header setup.

I finally went back to RTFM and I see there is a lot I could do with PWM settings. I have quite a hodgepodge of Thunderbird Turbocoupe injectors, and I wouldn't be surprised if at least some of them aren't jiving just right.

[TheShagg]

These ford thunderbird turbocoupe brown top injectors seem to like 1.3ms or more initial pulse. I tried to turn it higher but megatune advised that I shouldn't, so I stopped there. This may be part of why running rich makes it run better.

Now I need to track down why my RPM reading goes nuts at 3k RPM.

[Matt Cramer]

Yep, I don't have any crazy 360 degree header setup.

I finally went back to RTFM and I see there is a lot I could do with PWM settings. I have quite a hodgepodge of Thunderbird Turbocoupe injectors, and I wouldn't be surprised if at least some of them aren't jiving just right.

That is quite likely. I'd pull the batch, send it to Witch Hunter or a similar outfit, and get them all cleaned and flow tested.